Question AMD Rembrandt/Zen 3+ APU Speculation and Discussion

[tamz_msc]

Efficency has always been the main topic once we re talking of mobile devices, and in this case it s counter productive to use the CPU at full frequency on ST since it would exceed 15W and hence reduce battery life uselessly.

Now if you dont bother for this metric you can always set a power profile that would negate this limitation, but then the advertised run times would be no more accurate.

I'm not discussing efficiency. What you are saying is akin to saying that range is the main topic when discussing an EV. I want to talk about it's acceleration, not range.

Do you see the analogy?

 

[moinmoin]

You literally have the data right here posted in this very thread.

I literally have the supposed sluggishness on my lap right now, even on the older Renoir gen, and you're spouting pure hogwash calling it sluggish.

I'm not discussing efficiency. What you are saying is akin to saying that range is the main topic when discussing an EV. I want to talk about it's acceleration, not range.

Do you see the analogy?

Yeah, after all who cares about ranges of EVs?

 

[tamz_msc]

I literally have the supposed sluggishness on my lap right now, even on the older Renoir gen, and you're spouting pure hogwash calling it

Renoir drops as much as 40% in Geekbench while on battery.

Yeah, after all who cares about ranges of EVs?

One can care about both. But caring about one thing doesn't diminish the importance of the other.

 

[Abwx]

I'm not discussing efficiency. What you are saying is akin to saying that range is the main topic when discussing an EV. I want to talk about it's acceleration, not range.

Do you see the analogy?

His laptop use a 6850U, that is, a 15-28W SKU, on battery it s likely set to 15W for the reasons explained ad nauseam...

At 1361 pts ST GB it s barely slower than the 1521 pts when plugged, that s about 10% difference and not noticeable at all and moreover more efficient and better performing than any Intel SKU at this power, so what are you complaining about actually..?
That it s much better than your own laptop.??.

 

[moinmoin]

Renoir drops as much as 40% in Geekbench while on battery.

Not so surprising once you know that Geekbench has a lot of short benchmarks. Works as intended.

Better question would be if that behavior diminish performance in more steady workloads like e.g. games, and it doesn't.

And if one really doesn't like that delayed boosting behavior it's possible to turn it off (e.g. with ryzenadj).

One can care about both. But caring about one thing doesn't diminish the importance of the other.

Generally true. But we are in a thread about mobile, not desktop chips. Personally I'm not interested in DTR, or EVs with high acceleration but short range.

 

[tamz_msc]

His laptop use a 6850U, that is, a 15-28W SKU, on battery it s likely set to 15W for the reasons explained ad nauseam...

At 1361 pts ST GB it s barely slower than the 1521 pts when plugged, that s about 10% difference and not noticeable at all and moreover more efficient and better performing than any Intel SKU at this power, so what are you complaining about actually..?
That it s much better than your own laptop.??.

If it's 10% in Geekbench, it will be noticeable in other benchmarks that are more real-world oriented.

And I doubt that short duration power limit is 15 W on battery.

 

[tamz_msc]

Not so surprising once you know that Geekbench has a lot of short benchmarks. Works as intended.

Better question would be if that behavior diminish performance in more steady workloads like e.g. games, and it doesn't.

Nah, it means that AMD design goals is to sacrifice performance for battery life, even when told by the user not to.

Games are irrelevant to this discussion.

 

[moinmoin]

even when told by the user not to.

False. As I wrote users can turn it of if they want to. ryzenadj, and likely other tools I don't know can do that.

 

[tamz_msc]

False. As I wrote users can turn it of if they want to. ryzenadj, and likely other tools I don't know can do that.

I'm not talking about the delayed boost at all. That thing is absent from Cezanne onward.

 

[moinmoin]

I'm not talking about the delayed boost at all. That thing is absent from Cezanne onward.

What would cause your delta then? There doesn't appear to be different limits between battery and plugged in.

In any case ryzenadj allows switching between the built-in power-saving and max-performance modi that are otherwise automatically activated on battery and plugging in respectively, regardless of what that entails in the respective mobile Zen gen.

 

[tamz_msc]

What would cause your delta then? There doesn't appear to be different limits between battery and plugged in.

In any case ryzenadj allows switching between the built-in power-saving and max-performance modi that are otherwise automatically activated on battery and plugging in respectively, regardless of what that entails in the respective mobile Zen gen.

Again you fail to understand that boosting behavior doesn't only depend on power limits.

 

[jpiniero]

AMD Ryzen 7 7736U integrated Radeon 680M achieves Playstation 4 level graphics, runs God of War, Horizon Zero Dawn, and Final Fantasy XV at 1080p30

The Ryzen 7 7736U laptop would consume just 60 W when running games compared to 110 to 145 W on the PS4 while offering a similar level of performance. Integrated graphics have come a long way since the launch of the Sony console 10 years ago.

www.notebookcheck.net

Now there's a 7736U as well. Except this is technically slower than the 7735U. It's probably a straight rebrand of the 6800U while the 7735U is of the 6850U.

 

[LightningZ71]

Both 7735u and 7736u are listed on the AMD website. The 35 has a 50mhz advantage on boost clock with a fixed default tdp of 28w. The 36 has a configurable tdb from 15-28w. Both support the same memory speeds and the same 2200Mhz gpu clock. The 7736, according to the AMD site, is exactly a rebrand of the 6800u (jpiniero is correct). The 6850u also matches those specs. The Rembrandt "refresh" should just be named "rebrand" and any notable performance differences should be entirely related to tdp settings of the laptops and their thermal management capabilities. It would have been nice if AMD would have seen fit to "allow" an 8CU/4WGP configuration for the 6 core refresh, but that may be an exclusive Lenovo bin agreement for their Z series laptops.